RushTI
Run multiple TurboIntegrator processes simultaneously — with dependency scheduling, fault tolerance, and self-optimisation.
View on GitHub Read the DocsRun TI processes in parallel
TurboIntegrator processes typically run one at a time. RushTI removes that constraint, executing multiple processes simultaneously over a single TM1 connection — dramatically reducing total runtime for complex load cycles.
- Execute many TI processes concurrently with one connection
- Reduce load cycle runtimes by running independent processes in parallel
- Define execution order with simple task files — no code required
- Works with any IBM Planning Analytics / TM1 instance
Dependency-aware scheduling
Not all processes can run at the same time. RushTI uses a directed acyclic graph (DAG) to model dependencies between tasks, ensuring processes only run once their prerequisites are complete — in the correct order, every time.
- Define dependencies between processes declaratively
- DAG-based scheduling guarantees correct execution order
- Maximises parallelism while respecting data dependencies
- Visual dependency graphs for debugging and documentation
Fault tolerant and self-optimising
Long-running load cycles shouldn’t fail silently. RushTI includes checkpoint and resume capabilities so that a failed process doesn’t force a full restart, and self-optimisation continually improves scheduling based on past run times.
- Resume from the last successful checkpoint after a failure
- Detailed logging for every process execution
- Self-optimises scheduling order based on historical runtimes
- Configurable retry logic for transient failures
Free and open source
RushTI is built on TM1py and is free to use. The source is hosted on GitHub — contributions and bug reports from the community are welcome.
- Free to use with no licensing restrictions
- Built on TM1py — the Python framework for IBM Planning Analytics
- Actively maintained by Cubewise
- All code provided without warranty — use and modify freely